Bangla Text Summarization Using Encoder Decoder Model

Abstract

This time of information driven advancement has made robotized significant and significant information extraction a need. Computerized content synopsis has made it conceivable to extricate significant data from a lot of information without requiring any supervision. In any case, the extricated data could appear to be counterfeit on occasion and that is the place the abstractive synopsis strategy attempts to emulate the human method for outlining by making intelligent rundowns utilizing novel words and sentences. Because of the troublesome idea of this strategy, before profound learning, there hasn't been a lot of progress. In this way, during this work, we have proposed a consideration system-based grouping to-arrangement system to create abstractive outlines of Bengali content. We have likewise assembled our very own huge Bengali news dataset and applied our model on it to demonstrate for sure profound succession to-arrangement neural systems can accomplish great execution condensing Bengali writing